I am working on a simple pilot project comparing the capability of SQL,
SAS and R to perform a rolling mean per the following instructions. I
have completed the SQL and SAS analysis, so now it's R's turn.
Calculate mean values of x (x=count) for each date in the dataset where
mean = the average count of days [t-9] through day [t-3] for each
date/illness combination.
Dataset aggpilot
date illness x
1 2006/01/01 DERM 319
2 2006/01/02 DERM 388
3 2006/01/03 DERM 336
4 2006/01/04 DERM 255
5 2006/01/05 DERM 177
6 2006/01/06 DERM 377
7 2006/01/07 DERM 113
8 2006/01/08 DERM 253
9 2006/01/09 DERM 316
10 2006/01/10 DERM 187
11 2006/01/11 DERM 292
12 2006/01/12 DERM 275
13 2006/01/13 DERM 355
... 3102 total observations
Result of command: str(aggpilot)
'data.frame': 3102 obs. of 3 variables:
$ date : Factor w/ 1551 levels "2006/01/01","2006/01/02",..: 1 2 3 4
5 6 7 8 9 10 ...
$ illness: Factor w/ 2 levels "DERM","FEVER": 1 1 1 1 1 1 1 1 1 1 ...
$ x : int 319 388 336 255 177 377 113 253 316 187 ...
The results should look something like this:
DATE ILLNESS x MEAN NOTES [ROLLING MEAN DATE
RANGE]
1/1/2006 DERM 319
1/2/2006 DERM 388
1/3/2006 DERM 336
1/4/2006 DERM 255 319 [1/1/2006 -
1/1/2006]
1/5/2006 DERM 177 353.5 [1/1/2006 -
1/2/2006]
1/6/2006 DERM 377 347.6666667 [1/1/2006 -
1/3/2006]
1/7/2006 DERM 113 324.5 [1/1/2006 -
1/4/2006]
1/8/2006 DERM 253 295 [1/1/2006 -
1/5/2006]
1/9/2006 DERM 316 308.6666667 [1/1/2006 -
1/6/2006]
1/10/2006 DERM 187 280.7142857 [1/1/2006 -
1/7/2006]
1/11/2006 DERM 292 271.2857143 [1/2/2006 -
1/8/2006]
1/12/2006 DERM 275 261 [1/3/2006 -
1/9/2006]
1/13/2006 DERM 355 239.7142857 [1/4/2006 -
1/10/2006]
Ken Hall
Computer Scientist
Division of Healthcare Information (DHI) (proposed)
Public Health Surveillance Program Office (proposed)
Office of Surveillance, Epidemiology, & Laboratory Services (OSELS)
(proposed)
Centers for Disease Control & Prevention (CDC)
k...@cdc.gov
Mobile: 404-993-3311
Office: 404-498-6839
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
